Skybuck's Dream PC, Third Build, Done 99.9% ;) (Pictures!)

I like this photo:

Lots of band-aids. Perhaps using a big sheet of sticky back kitchen shelf paper would have been more useful? Other than increase the capacitance to ground, what are all the band-aids suppose to do? The wires aren't going to fall out. Nobody is going to see it with the covers on. The glue will eventually make a sticky mess.

Nice table, but far too easy to scratch. I suggest bigger and more cardboard to protect it.

Also, I note a #1 Philips screwdriver. That's good for some screws, but most PC screws require a #2 Philips or 1/4" driver.

Hmmm....

Lots of air going nowhere. Looks like air goes in through the two fans on the back. It goes to the CPU base and blown upward by the CPU fan. If the power supply fan ran backwards, the hot air would go out through the power supply. However, the power supply fan also seems to be blowing inward, which isn't going to work. First, the hot air wants to rise. Second, there's lots of intakes, but no exhaust. Note that the exhaust port area has to be somewhat larger than the intake due to the expansion of hot hair. Get and IR thermometer and see if this mess is going to work. If not, leave the PS fan alone and reverse the flow of the two rear fans so that the hot air goes OUT of the box, not in.

Now THAT'S funny!

Chris

I doubt seriously that the PS fan blows into the case. I believe that violates the ATX PSU spec.

It's retarded. No... It's beyond retardation.

Yes. I have never once found myself standing in awe of a non-functional computer.

Assumption, the mother of all screwups.

Look at the photograph:

Which way would you assume that the PS supply fan is going to rotate? My astute guess, is that with the swept back blade configuration, the direction of rotation is clockwise. If that's true, then the fan is moving air INTO the case. I'll admit that it's possible that I'm wrong, which is why I asked Mr Flying to verify the air flow and rotation.

Y'er wrong about the ATX specs. ATX Motherboard Spec 2.01:

See section 4.1 which proclaims: The intended location and fan direction in an ATX system is for the power supply fan to draw in cool air from outside the chassis and exhaust it directly onto the processor.

Never mind that this "cool" outside air is pre-heated by the power supply before it gets to the processor and fights convective air flow (hot air like to rise).

I once wrote a rather nasty critique of the air flow problems with the ATX design when it first appeared. I couldn't find my original, but this rant from Dec 2000 covers the main points:

Weird. The Intel web pile has version 2.01. The current version is apparently 2.2. See:

The aformentioned hint as to airflow direction was deleted and replaced by a vague section 5.1 (Venting) that fails to specify the air flow direction. The power supply should be placed in close proximity to the processor if the power supply is expected to cool the processor properly... Not very useful or specific. I guess the airflow is the responsibility of the builder.

Except that it is not an assumption. So it appears to be the mother of your f*ck up.

Absolutely negligible in a 2 foot tall case. In a rack... maybe a percent or so.

Jeff Liebermann wrote in news: snipped-for-privacy@4ax.com:

That means the heat from the power supply gets added to the heat from the microprocessor.Not very good,IMO.

Then what is the microprocessor(uP) heatsink cooling fan for? If you examine the airflow on that,it draws air thru the fan and forces it thru the uP heatsink.Having a PS fan blow warmed air at the HS isn't going to help uP cooling any. uP/HS airflow will push away that PS airflow.

DC fans generally blow air toward the side that has the spokes holding the fan motor in place.

Or another way of looking at it, fans are designed to blow towards the concave side of the spokes because thats why the spokes are curved in the first place. eg if one is fanning oneself with a magazine which way you naturally curve it?

Lordy

I see a lot of ATX computers that have a shroud over the CPU heatsink, and use the PS fan to pull the heat away. Quite often, they have bad electrolytics in the CPU power supply.

PS supply fan? Power supply fan? There's no such thing on that picture :) The PSU is at the bottom of the box (Antec P1xx I think)

Look:

Yep. See my 8 year old rant on the subject at:

There are additional problems. There's also the problem of the air intake/exhaust port sizing. Basically, Intel had the grand idea of using one fan to cool both the PS and the CPU. That might have worked with the PIII series of CPU's, which only burned about 60 watts of heat. Enter the various P4 mutations, and they'll burn from 60 to 150 watts:

Meanwhile the efficient of the power supplies run about 75% which adds another 25 to 40 watts of heat directly onto the CPU. I vaguely recall that Intel got some kind of design award for the ATX abomination.

In this case, the problem with Mr Flying's overheating machine is best solved by simply changing the direction of the case fans to blow air out of the box. It's not optimum, but is easier than reversing the CPU fan.

Even Intel includes a CPU heatsink and fan on their motherboards. The documentation that comes with P4 processors, warns of dire consequences if the required heatsink and CPU fan are not properly sillycone greased and securely mounted. I guess the guys that did the CPU and motherboards didn't read the ATX guidelines.

Yep. Worse, if you measure the exhaust port diameter, which is the area between the heatsink fins, it's considerably less than the intake area. That means the fan is fighting back pressure. It would work better if the fan blew air OUT of the heat sink, but that's not going to happen. The best compromise I've seen are two fans, mounted vertically, at each end of the heat sink fins. I have one of those on my office machine, with a carboard duct pointing directly out the back of the machine. I haven't done much testing to see if it's really any better than the conventional fans, but the CPU seems to be running cooler.

You might be right. It sure looks like a power supply to me, but I can't see any components through the grill work, so it might be something else. Mounting the PS at the bottom of the box is the right answer for hot air rising. However, I still contend that the upper (PS?) fan in the picture blows air into the box, as do the 2 fans mounted on the back panel. No way to be sure about the air flow until Mr Flying recovers his sanity and discloses the direction. However, if my guess(tm) is correct, and the bottom mounted power supply is blowing air into the box, then both the rear fans, and the mystery upper fan need to be blowing air OUT of the box.

Incidentally, bottom mounted intake fans tend to suck air, dust, dirt, filth, crud, and whatever is on the floor, carpet, or table that they're placed upon. I like to mount such boxes on some manner of spacers go get the box off the floor.

Fans blow out from concave side of fins.

Lordy